Rate Transient Analysis

Toll Free: 1.800.625.2488 :: Phone: 403.213.4200 :: Email: fast@fekete.
com
fekete.com
Modern Production Data Analysis
Day 1 - Theory
1. Introduction to Well Performance Analysis
4. Theory of Type Curves
2. Arps – Theory a) Dimensionless variables
a) Exponential b) The log-log plot
b) Hyperbolic c) Type Curve matching
c) Harmonic
5. Principle of Superposition
3. Analytical Solutions a) Superposition
a) Transient versus Boundary Dominated b) Desuperposition
Flow c) Material Balance Time
b) Boundary Dominated Flow
i. Material Balance Equation 6. Gas Corrections
ii. Pseudo Steady-State Concept a) Pseudo-Pressure
iii. Rate Equations b) Pseudo-Time
c) Transient Flow
i. Radius of Investigation Concept
ii. Transient Equation (Radial Flow)
Modern Production Data Analysis
Day 2 - Practice
11. Modeling and History Matching
7. Arps – Practical Considerations
a) Guidelines
12. A Systematic and Comprehensive
b) Advantages
Approach
c) Limitations
13. Practical Diagnostics
8. Analysis Using Type Curves
a) Data validation
a) Fetkovich
b) Pressure support
b) Blasingame (Integrals)
c) Interference
c) AG and NPI (Derivatives)
d) Liquid loading
d) Transient
e) Accumulating skin
e) Wattenbarger
damage
f) Transient flow regimes
9. Flowing Material Balance
14. Tutorials
10. Specialized
15. Selected Topics and Examples
Introduction to Well
Performance Analysis
Traditional
- Production rate only
- Using historical trends to predict future
- Empirical (curve fitting)
- Based on analogy
- Deliverables:
- Production forecast
- Recoverable Reserves under current conditions
Modern
- Rates AND Flowing Pressures
- Based on physics, not empirical
- Reservoir signal extraction and characterization
- Deliverables:
- OGIP / OOIP and Reserves
- Permeability and skin
- Drainage area and shape
- Production optimization screening
- Infill potential
Recommended Approach
- Use BOTH Traditional and Modern together
- Production Data Analysis should include a

comparison of multiple methods
- No single method always works
- Production data is varied in frequency, quality

and duration
Modern Production Analysis -
Integration of Knowledge
Modern Production Analysis
Welltest Analysis
Empirical Decline
Analysis
- Flow regime
- High resolution characterization over
early-time - Characterization life of well
characterization of perm and skin
- Projection
- Estimation of fluids-
of recovery - Estimation of
- High resolution -Estimation of in-place
constrained reserves when
characterization contacted flowing pressure
by historical
of the near- drainage area - Performance based
operating is unknown
wellbore recovery factor
conditions
-Estimation of
-Point-in-time reservoir - Able to analyze
characterization pressure transient production
of wellbore skin data (early-time
production, tight gas
etc)
Arps - Empirical
Traditional Decline Curves
– J.J. Arps
- Graphical – Curve fitting exercise
- Empirical – No theoretical basis
- Implicitly assumes constant operating conditions

The Exponential Decline Curve
Unnam e d We ll Rate vs Time
5.00
4.50
4.00
q  qie  Dit
3.50
3.00
Gas Rate, MMscfd
Slope
2.50
2.00
Di 
1.50
1.00
q
0.50
0.00
2001 2002 2003 2004 2005 2006
Unnam ed Well Rate vs Time

Unnamed Well Rate vs. Cumulative Prod.
101
4.50
7
6
Dit
log q  log qi 
5 4.00
q  qi  DiQ
4
2.302
3.50
3
Di  Slope
3.00
Di  2.302* Slope
MMscfd
MMscfd
2.50
1.0
Gas Rate,
Gas Rate,
2.00
7
6
5
1.50
4
3
1.00
2
0.50
0.00
10-1
0.00 0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 0.90 1.00 1.10 1.20 1.30 1.40 1.50 1.60 1.70 1.80 1.90 2.00 2.10 2.20 2.30 2.40 2.50
2001 2002 2003 2004 2005 2006
Gas Cum. Prod., Bscf
The Hyperbolic Decline Curve
Unnam ed Well Rate vs. Cumulative Prod.
4.50
4.00
3.50
qi
q
(1  bDit )1/ b
3.00
MMscfd
2.50
Di b
D b q
Gas Rate,
2.00
1.50
qi
1.00
0.50
D  f (t )
0.00
0.00 0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 0.90 1.00 1.10 1.20 1.30 1.40 1.50 1.60 1.70 1.80 1.90 2.00 2.10 2.20 2.30 2.40 2.50 2.60
Gas Cum . Prod., Bscf

Hyperbolic Exponent “b”
Unnam e d We ll Rate vs. Cumulative Prod.
4.50
4.00
3.50
3.00
Mild Hyperbolic – b ~ 0
Gas Rate, MMscfd
2.50
2.00
1.50
1.00
0.50
0.00
0.00 0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 0.90 1.00 1.10 1.20 1.30 1.40 1.50 1.60 1.70 1.80 1.90 2.00 2.10 2.20 2.30 2.40 2.50 2.60
Gas Cum . Pr od., Bs cf
NBU 921-22G Rate vs. Cumulative Prod.
3.20
3.00
Strong Hyperbolic – b ~ 1
2.80
2.60
2.40
2.20
2.00
Gas Rate, MMscfd
1.80
1.60
1.40
1.20
1.00
0.80
0.60
0.40
0.20
0.00
0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40 0.45 0.50 0.55 0.60 0.65 0.70 0.75 0.80 0.85 0.90 0.95 1.00 1.05
Gas Cum ulative , Bs cf

Analytical Solutions
Transient vs Boundary Dominated Flow
Transient Flow
- Early-time OR Low Permeability
- Flow that occurs while a pressure “pulse” is

moving out into an infinite or semi-infinite acting
reservoir
- Like the “fingerprint” of the reservoir

- Contains information about reservoir
properties (permeability, drainage shape)
Boundary Dominated Flow
- Late-time flow behavior
- Typically dominates long-term production data
- Reservoir is in a state of pseudo-equilibrium –

physics reduces to a mass balance
- Contains information about reservoir pore volume

(OOIP and OGIP)
Definition of Compressibility
pi pi-dp
dV
V V
1 V
c
V p
Compressibility Defines Material Balance of a
Closed Oil Reservoir (above bubble point)
Dp = pi - p DV = Np
V=N
1 Np
c
N pi  p
Np
p  pi 
ctN
p  pi  mpssNp
Note: only valid if c is constant

Single Phase Oil MB
pi  p y  mx
pi  p  mpssNp
slope  mpss
Np
Illustration of Pseudo-Steady-State
p1
1
2 p2
pressure
3 p3
pwf1
pwf2 Constant Rate q

time
pwf3
rw Distance re
Flowing Material Balance
y  mx  b
pi  pwf  mpssNp  b
pi  pwf
slope  mpss
Np
Steady-State Inflow Equation
pi
p
pressure
p  pwf  qbpss
bpss  f (kh, s, area)
pwf
Inflow (Darcy) pressure drop- Constant-

Productivity Index
rw Distance re
Variable Rate
y  mx  b
pi  pwf mpssNp
  bpss
pi  pwf q q
q
slope  mpss
bpss
Np
q
The Three Most Important Equations
in Modern Production Analysis
p  pi  mpssNp
p  pwf  qbpss
pi  pwf  mpssNp  qbpss

Operating Conditions - Simplified
Constant Pressure Constant Rate
= =
Production Welltest
q q
pwf pwf
Constant Rate Solution
Relate Back to Arps Harmonic
- Invert the PSS equation
q 1 1
 
pi  pwf (t ) mpssNp  bpss mpsst  bpss
q
1
q bpss

pi  pwf (t ) mpss t  1
bpss
Constant Flowing Pressure Solution
- Required: q(t), Npmax and N for constant pwf
- Take derivative of both equations and solve for q
- Integrate to find Np(t), as t goes to infinity Np goes to

Npmax
pi  pwf
mpss
 t
q(t )  e bpss
bpss
pi  pwf
Np max    pi  pwf  ctN
mpss
Constant Flowing Pressure Solution
Relate Back to Arps Exponential, Determine N
pi  pwf
qi 
bpss
mpss
Di 
bpss
qi
Np max 
Di
ct ( pi  pwf ) ct ( pi  pwf ) Di
N 
Np max qi
Plot Constant p and Constant q together
1
0. 9
Constant rate q/Dp (Harmonic)

0. 8
1
q bpss

0. 7
pi  pwf (t ) mpss t  1
0. 6
bpss
0. 5
0. 4
0. 3
Constant pressure q/Dp (Exponential)

0. 2
mpss
q(t ) 1  bpss t
 e
pi  pwf bpss
0. 1
0 5 10 15 20 25 30 35 40 45
Transient Flow
Transient and Boundary Dominated Flow
Numerical Radial Model
10 Cross Section Pressure Plot
3600
3400
3200
3000
Cross Section
2800
2600
Transient Well Boundary Dominated
2400
Performance = f(k, skin, Well Performance =
2200
time) f(Volume, PI)
psi
2000
Pressure,
1800
1600
1400
1200
1000
Plan View 800
600
400
200
0
-4000 -3600 -3200 -2800 -2400 -2000 -1600 -1200 -800 -400 0 400 800 1200 1600 2000 2400 2800 3200 3600 4000
Radii, ft
Radius (Region) of Investigation
Numerical Radial Model
10 Cross Section Pressure Plot
3600
3400
3200
kt
Cross Section 3000
rinv 
2800 948 c
 kt
2600
2400 Ainv 
2200 948 c
psi
2000
Pressure,
1800
1600
1400
1200
1000
Plan View 800
600
400
200
0
-4000 -3600 -3200 -2800 -2400 -2000 -1600 -1200 -800 -400 0 400 800 1200 1600 2000 2400 2800 3200 3600 4000
Radii, ft
Transient Equation
Describes radial flow in an infinite acting reservoir
q kh 1

( pi  pwf ) 141.2 B 1  0.0063kt 
ln    0.4045  s
2   ct 
q(t)’s compared
1. 6
1. 4
1. 2
Transient flow: compares to Arps “super
1
hyperbolic” (b>1)
0. 8
0. 6
0. 4
0. 2
0 5 10 15 20 25 30 35 40 45
Type Curves
Blending of Transient into
3
Complete q(t) consists of:

2. 5 Transient q(t) from t=0 to tpss
Depletion equation from t = tpss and higher
2
1. 5
0. 5
0 5 10 15 20 25 30 35 40 45
Log-Log Plot: Adds a New
Visual Dynamic
Comparison of qD with 1/pD
Cylindrical Reservoir with Vertical Well in Center
1000
Infinite Acting Boundary Dominated

100

10 Harmonic 0.9
1
qD and 1/pD
0.1
0.01
0.001
Constant Pressure Solution Exponential
0.0001
0.00001
0.000001
0.000001 0.0001 0.01 1 100 10000 1000000 100000000 1E+10 1E+12 1E+14
tD
Type Curve
- Dimensionless model for reservoir / well system
- Log-log plot
- Assumes constant operating conditions
- Valuable tool for interpretation of production and

pressure data
Type Curve Example - Fetkovich
Fetkovich Typecurve Analysis
1.0
7 Harmonic
6
5
1
4
q (t ) qDd 
3 qDd  1  tDd
qi
2
tDd  Dit
qDd
Rate,
10-1
9
Exponential Hyperbolic
qDd  e tDd
7
6
1
qDd 
5
4
(1  btDd )1/ b
3
10-2
2 3 4 5 6 7 8 9 2 3 4 5 6 7 8 2 3 4 5 6 7 8
10-1 1.0
tDd Tim e
101
Plotting Fetkovich Type Curves-
Example
Time (years) Rate (MMscfd) tDd qDd
Well 1 (exponential) Well 1 Well 2 Well 1 Well 2 Well 1 Well 2
0 2.50 10.00 0.00 0.00 1.00 1.00
qi = 2.5 MMscfd
1 2.26 8.19 0.10 0.20 0.90 0.82
Di = 10 % per year 2 2.05 6.70 0.20 0.40 0.82 0.67
3 1.85 5.49 0.30 0.60 0.74 0.55
Well 2 (exponential) 4 1.68 4.49 0.40 0.80 0.67 0.45
5 1.52 3.68 0.50 1.00 0.61 0.37
qi = 10 MMscfd 6 1.37 3.01 0.60 1.20 0.55 0.30
Di = 20 % per year 7 1.24 2.47 0.70 1.40 0.50 0.25
8 1.12 2.02 0.80 1.60 0.45 0.20
9 1.02 1.65 0.90 1.80 0.41 0.17
10 0.92 1.35 1.00 2.00 0.37 0.14
Raw Data Plot Dimensionless Plot
12.00 1.00
10.00
Rate (MMscfd)
8.00
Well 1
qDd
Well 1
6.00
Well 2 Well 2
4.00
2.00
0.00 0.10
0 5 10 15 0.01 0.10 1.00 10.00
Time (years) tDd
Fetkovich Typecurve Matching
In most cases, we don’t know what “qi” and “Di” are ahead of time. Thus, qi and Di
are calculated based on the typecurve match (ie. The typecurve is superimposed on
the data set
NBU 921-22G Fetkovich Typecurve Analysis
1.0
q (t ) 7
qi 
q6
5
qDd Rate, 4
tDd
Di 
2
t
qDd
t 10-1
9
8
3 4 5 6 7 8 9
1.0
tDd 2 3 4 5 6 7 8 9
101
2
Tim e
Knowing qi and Di, EUR (expected ultimate recovery) can be calculated

Analytical Model Type Curve
Fetkovich Typecurve Analysis
101
4
3
1.0
9
6 Transient Flow
4
3
2
Rate,
10-1
9
re/rwa = 10 re/rwa = 100 re/rwa = 10,000
qDd 6
4
3
10-2
9
4
3 Boundary Dominated Flow
2
Exponential
10-4
2 3 4 567 9
10-3
2 3 4 5 6 78
10-2
2 3 4 5 6 78
10-1
Tim e
tDd
2 3 4 5 6 78
1.0
2 3 4 5 6 78
101
2 3 4 5 67
Modeling Skin using Apparent Wellbore
Radius
rwa (s)
s
ΔP(s) rwa  rwe
rwa(d)
ΔP(d)
rw re
Dimensionless Variable Definitions
(Fetkovich)
141.2q  B   re  1 
qDd   ln   
kh( pi  pwf )   rwa  2 
0.00634kt
 ctrwa 2
tDd 

1   r e  1   re 
2

 ln        1
2   rwa  2   rwa  
Type Curve Matching (Fetkovich)
The Fetkovich analytical typecurves can be used to calculate three parameters:
permeability, skin and reservoir radius
141.2 B   re  1  q
k  ln   
h( pi  pwf )   rwa  2  qDd match
0.00634k 1 t  rw 
rwa  s  ln  
 ct 1   re  1   re 2  tDd  rwa 
 ln        1
2   rwa  2   rwa   match
141.2 B 0.00634 q t
re  2
h( pi  pwf )  ct qDd match tDd match
Type Curve Matching - Example
10 Fetkovich Typecurve Analysis
101
8
6
k = f(q/qDd)
4
3 reD = 50 s = f(q/qDd * t/tDd, reD)
2 re = f(q/qDd * t/tDd)
1.0
8
6
4
q Transient Flow
3
2
Rate,
10-1
8
qDd 6
4
3
10-2
t
8
6
4
3
2
Exponential
10-3
2 3 4 5 6 78 2 3 4 5 6 78 2 3 4 5 6 78 2 3 4 5 6 78 2 3 4 5 6 78 2 3 4 5 6 78
10-4 10-3 10-2 10-1
Tim e
tDd 1.0 101
Superposition
What about Variable Rate / Variable Pressure
Production? The Principle of Superposition
Superposition in Time:
1. Divide the production history into a series of constant rate periods

2. The observed pressure response is a result of the additive effect of each rate
change in the history
Example: Two Rate History
q2
q q1
pi  pwf  q1 f (t )  (q 2  q1) f (t  t1)
pwf
Effect of (q2-q1)
t1
The Principle of Superposition
Two Rate History
pi  pwf  q1 f (t )  (q 2  q1) f (t  t1)
N - Rate History
N
pi  pwf   (qj  qj  1) f (t  tj  1)
j 1
f(t) is the Unit Step Response
Superposition versus Desuperposition
Simple Complex
- Unit step response f(t) - Real rate and pressure

- Type Curve history
- Superposition Time Superposition - Modeling (history
matching)
q
q
pwf
Desuperposition
pwf
Superposition Time
Convert multiple rate history into an equivalent single rate history by re-plotting
data points at their “superposed” times
pi  pwf N
(qj  qj  1)
 f (t  tj  1)
qN j 1 qN
The Principle of Superposition –
PSS Case
pi  pwf N
(qj  qj  1)
 f (t  tj  1)
qN j 1 qN
pi  pwf t 141.2B  re 3 
f (t )     ln  
q ctN kh  rwa 4 
pi  pwf 1 N
(qj  qj  1) 141.2B  re 3 
qN

ctN

j 1 qN
(t  tj  1)   ln  
kh  rwa 4 
pi  pwf 1 Np 141.2B  re 3 
   ln  
qN ctN qN kh  rwa 4 
Superposition Time: Material Balance Time

Definition of Material Balance Time
(Blasingame et al)
Actual Rate Decline Equivalent Constant Rate
q
Q Q
actual material
time (t) balance = Q/q
time (tc)
Features of Material Balance Time
-MBT is a superposition time function
- MBT converts VARIABLE RATE data into an

EQUIVALENT CONSTANT RATE solution.
- MBT is RIGOROUS for the BOUNDARY

DOMINATED flow regime
- MBT works very well for transient data also, but

is only an approximation (errors can be up to 20%
for linear flow)
MBT Shifts Constant Pressure to
Equivalent Constant Rate
Comparison of qD (Material Balance Time Corrected) with 1/pD
Cylindrical Reservoir with Vertical Well in Center
1000 1.2
Very early time radial flow
Ratio (qD to 1/pD) ~ 90%
100
1
0.97
10

1 0.8
1/pD
Ratio 1/pD to qD
Harmonic
qD and 1/pD
0.1
Beginning of "semi-log" radial flow (tD=25) 0.6
Ratio (qD to 1/pD) ~ 97%
0.01
0.001 0.4
Constant Pressure Solution qD

0.0001
Corrected to Harmonic
0.2
0.00001
0.000001 0
0.000001 0.0001 0.01 1 100 10000 1000000 100000000 1E+10 1E+12 1E+14
tD
Corrections for Gas Reservoirs
Corrections Required for Gas
Reservoirs
• Gas properties vary with pressure
– Formation Volume Factor
– Compressibility
– Viscosity
Corrections Required for Gas
Reservoirs
Depletion Term Reservoir FlowTerm:

Depends on Depends on “B” and
compressibility Viscosity
qt 141.2qBo  re 3 
pi  pwf    ln  
coN kh  rwa 4 
Darcy’s Law Correction for Gas
Reservoirs
Darcy’s Law states : Dp  q
For Gas Flow, this is not true because

viscosity () and Z-factor (Z) vary with pressure
Solution: Pseudo-Pressure
p
pdp
pp  2
0
Z
Depletion Correction for Gas
Reservoirs
Gas properties (compressibility and viscosity) vary
significantly with pressure
Gas Compressibility
0.012
0.01
0.008
Compressibility (1/psi)
0.006
1
cg 
0.004
p
0.002
0
0 1000 2000 3000 4000 5000 6000
Pressure (psi)
Depletion Correction for Gas
Reservoirs: Pseudo-Time
Solution: Pseudo-Time
ta  cg i 
dtt
0 c g
 , c g  Evaluated
pressure
at average reservoir
Not to be confused with welltest pseudo-time which evaluates properties

at well flowing pressure
Equation for Gas
Constant Rate Case
Pseudo-pressure Pseudo-time
2 pi 1.417e6 * Tq  re 3 
Dpp  ppi  ppwf  qta   ln  
( cgZ )iGi kh  rwa 4 
Variable Rate Case

Pseudo-Cumulative Production
Dpp Gpa
  bpss
q qGi
Overall time function - Material
Balance Pseudo-time
1 t
tc   qdt
q 0
1 ta
tca   qdta 
cg i t qdt
q 0 q 0  c g
Improved Material Balance
Pseudo-time
Overall material balance pseudo-time function (corrected for
variable fluid saturations, water encroachment, in-situ fluids & formation expansion and
desorption):
ct i t q(t )
tca  
q 0 c t 1  cf ( pi  p )
dt
Arps – Practical Consideration
Notes About Drive Mechanism and
b Value (from Arps and Fetkovich)
b value Reservoir Drive Mechanism
0 Single phase liquid expansion (oil above bubble point)

Single phase gas expansion at high pressure
Water or gas breakthrough in an oil well
0.1 - 0.4 Solution gas drive
0.4 - 0.5 Single phase gas expansion
0.5 Effective edge water drive
0.5 - 1.0 Layered reservoirs
>1 Transient (Tight Gas)

Advantages of Traditional
- Easy and convenient
- No simplifying assumptions are required regarding the

physics of fluid flow. Thus, can be used to model very
complex systems
- Very “Real” indication of well performance

Limitations of Traditional
- Implicitly assumes constant operating conditions
- Non-unique results, especially for tight gas (transient flow)
- Provides limited information about the reservoir

Example 1: Decline Overpredicts
Reserves
Unnam ed Well Rate vs Time
MMscfd
Gas Rate,
October Novem ber Decem ber January February March April

2001 2002
EUR = 9.5 bcf

MMscfd
3
Gas Rate,
0
0.00 0.50 1.00 1.50 2.00 2.50 3.00 3.50 4.00 4.50 5.00 5.50 6.00 6.50 7.00 7.50 8.00 8.50 9.00 9.50 10.00 10.50

Example 1 (cont’d)
Flowing Pressure and Rate vs Cumulative Production
Rates
5 1200
4.5
1000
4
Flowing Pressure (psia)

3.5 True EUR does not
800
Rate (MMscfd)
exceed 4.5 bcf

3
2.5 Pressures 600
2 Forecast is not
valid here 400
1.5
1
200
0.5
0 0
0 1 2 3 4 5 6 7 8 9 10
Cumulative Production (bcf)
Example 2: Decline Underpredicts
Reserves
8.50
8.00
7.50
7.00
6.50
6.00
EUR = 3.0 bcf

5.50
5.00
MMscfd
4.50
Gas Rate,
4.00
3.50
3.00
2.50
2.00
1.50
1.00
0.50
0.00
0.00 0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 0.90 1.00 1.10 1.20 1.30 1.40 1.50 1.60 1.70 1.80 1.90 2.00 2.10 2.20 2.30 2.40 2.50 2.60 2.70 2.80 2.90 3.00 3.10 3.20

Unnam ed Well Flowing Material Balance
0.085
Legend
Decline FMB
0.080
0.075
0.070
0.065
0.060
OGIP = 24 bcf
6 psi 2 /cP)
0.055
0.050
MMscfd/(10
0.045
0.040
Normalized Rate,
0.035
0.030
0.025
0.020
0.015
0.010
Original Gas In Place

0.005
0.000
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
Norm alized Cum ulative Production, Bscf

Unnam ed Well Data Chart
18
Legend 1300
Pressure
17 Actual Gas Data
1200
16
15 1100
14
1000
13
900
12
11 800
Operating conditions: Low drawdown
Pressure,
10
Gas, MMscfd
700
9
Increasing back pressure
psi
600
8
7
500
6
400
5
4 300
3
200
100
1
0 0
0 20 40 60 80 100 120 140 160 180 200 220 240 260 280 300 320 340 360 380 400 420 440 460 480 500 520 540 560 580 600 620 640 660 680 700 720
Tim e, days
Example 3 – Illustration of Non-
Uniqueness
Arps Production Forecast
10
1
Gas Rate (MMscfd)
0.1
Economic Limit =
0.05 MMscfd b = 0.25, b = 0.50, b = 0.80,
EUR = 2.0 bcf EUR = 2.5 bcf EUR = 3.6 bcf
0.01
Dec-00 May-06 Nov-11 May-17 Oct-22 Apr-28 Oct-33
Time
Analysis using Type Curves
Blasingame Typecurve Analysis
Blasingame typecurves have identical format to those of Fetkovich. However, there

are three important differences in presentation:
1. Models are based on constant RATE solution instead of

constant pressure
2. Exponential and Hyperbolic stems are absent, only

HARMONIC stem is plotted
3. Rate Integral and Rate Integral - Derivative typecurves are used

(simultaneous typecurve match)
Data plotted on Blasingame typecurves makes use of MODERN DECLINE ANALYSIS

methods:
- NORMALIZED RATE (q/Dp)
- MATERIAL BALANCE TIME / PSEUDO TIME

Blasingame Typecurve Analysis-
Comparison to Fetkovich
Fetkovich Blasingame
log(q) log(q/Dp)
log(qDd) log(qDd)
log(t) log(tca)
log(tDd) log(tDd)
- Usage of q/Dp and tca allow boundary dominated flow to be represented by harmonic
stem only, regardless of flowing conditions
- Blasingame harmonic stem offers an ANALYTICAL fluids-in-place solution
- Transient stems (not shown) are similar to Fetkovich

Definitions
Typecurves Data - Oil Data - Gas

141.2q   re  1  q q
Normalized Rate qDd  ln rwa   2 
khDP     DP DPp
t DA
 q  1 q
t
c
 q  t ca
 q t dt
Rate Integral 1
      1 q
q Ddi 
t DA
Dd 
 DP i tc 0 DP
dt
 DP  t  DP dt
0  p i ca 0 p
Rate Integral - Derivative
q Ddid  t DA
dq Ddi  q   q 
d  tca d  
dt DA  q   DP i  q   DP 
   tc    p i
 DP id 
dtc  DP  dtca
 p id
Concept of Rate Integral
(Blasingame et al)
actual rate rate
integral =
Q/t
Q Q
actual actual
time time
Rate Integral: Like a Cumulative
Average
Average rate over time period
“0 to t1”
Average rate over time period
“0 to t2”
t1 t2
Effective way to remove noise

Rate Integral: Definition
 q  1 q
tc
    dt
 Dp i tc 0 Dp
Typecurve Interpretation Aids:
Integrals, Derivatives
Typecurve Most Useful For Drawback Used in Analysis
Integral / Removing the scatter from Dilutes the reservoir Fetkovich,

Cumulative noisy data sets signal Blasingame, NPI
Amplifying the reservoir

Amplifies noise - Agarwal-Gardner,
Derivative signal embedded in
often unusable PTA
production data
Maximizing the strengths

Integral-Derivative Can still be noisy Blasingame, NPI
of Integral and Derivative
Other methods: Data filtering, Moving averages, Wavelet decomposition

Rate Integral and Rate Integral
Derivative (Blasingame et al)
Rate Integral
Rate (Normalized)
Rate Integral Derivative

Transient Calculations
Oil:
k is obtained from rearranging the definition of
q  141.2    re  1
q Dd    ln    
Dp  kh   rwa 
 match 2 
q 
 Dp   141.2    re  1
k     ln    
 q Dd     rwa 
h  match 2 
  match
Solve for rwa from the definition of
0.006328 ktc
t Dd 

2   re 
2
  r  1
1
ct rwa     1  ln  e   
    rwa 
  wa  match  match 2 
2 r

 t 
  c  0.006328 k
r
wa t   2   
 Dd  match 1   re   r  1
c   1 ln  e   
t 2  r    rwa  2
  wa  match
  match 
r 
s  ln  w 
 rwa 
Boundary Dominated Calculations-Oil
Oil-in-Place calculation is based on the harmonic stem of Fetkovich typecurves.
In Blasingame typecurve analysis, qDd and tDd are defined as follows:
qDd 
q / Dp  and tDd  Ditc
q / Dp i
Recall the Fetkovich definition for the harmonic typecurve and the PSS equation for oil in
harmonic form: Definition of Harmonic PSS equation for oil in
typecurve
1 harmonic form, using
material balance time
1 q b
qDd  and 
1  tDd Dp 1
tc  1
From the above equations:
ctNb



q 

Dp  i  q  1
q
 
where    , and Di 
1
Dp 1  Ditc  Dp  b ctNb
 i
Boundary Dominated Calculations-Oil
Oil-in-Place (N) is calculated as follows:
Rearranging the equation for Di:
1
N
ctDib
Now, substitute the definitions of qDd and tDd back into the above equation:
1 1  tc   q / Dp 
N    
 tDd   qDd  ct  tDd   qDd 
ct    
 tc   q / Dp 
X-axis “match-point from Y-axis “match-point”
typecurve analysis from typecurve analysis
Blasingame Typecurve Analysis- Boundary
Dominated Calculations- Gas
Gas-in-Place calculation is similar to that of oil, with the additional complications of pseudo-
time and pseudo-pressure.
In Blasingame typecurve analysis, qDd and tDd are defined as follows:
qDd 
q / Dpp  and tDd  Ditca
q / Dpp i
Recall the Fetkovich definition for the harmonic typecurve and the PSS equation for gas in
harmonic form:
Definition of Harmonic PSS equation for gas in
typecurve harmonic form, using
1 material balance pseudo-
1 q b time
qDd  and 
1  tDd Dpp 2 pi
tca  1
From the above equations:
Zct iGib



q 

Dp  i  q  1
q
 
where    , and Di 
2 pi
Dp 1  Ditc  Dpp  b
 i
Zct iGib
Boundary Dominated Calculations- Gas
Gas-in-Place (Gi) is calculated as follows:
Rearranging the equation for Di:
2 pi
Gi 
Di Zct ib
Now, substitute the definitions of qDd and tDd back into the above equation:
2 pi 2 pi  tca   q / Dpp 
Gi  
 tDd   
Zct i  tDd   qDd 
 Zct i 
 q Dd 

 tca   ( q / Dpp ) 
 
X-axis “match-point from Y-axis “match-point”

typecurve analysis from typecurve analysis
Agarwal-Gardner Typecurve Analysis
Agarwal and Gardner have developed several different diagnostic

methods, each based on modern decline analysis theory. The AG
typecurves are all derived using the WELLTESTING definitions of
dimensionless rate and time (as opposed to the Fetkovich
definitions). The models are all based on the constant RATE
solution. The methods they present are as follows:
1. Rate vs. Time typecurves (tD and tDA format)
2. Cumulative Production vs. Time typecurves (tD and tDA

format)
3. Rate vs. Cumulative Production typecurves (tDA

format)
- linear format
- logarithmic format
Agarwal-Gardner Typecurve Analysis
Agarwal-Gardner - Rate vs.
Time Typecurves
Agarwal and Gardner Rate vs. Time typecurves are the same as
conventional drawdown typecurves, but are inverted and plotted in
tDA (time based on area) format.
qD vs tDA
The AG derivative plot is not a rate derivative (as per Blasingame).

Rather, it is an INVERSE PRESSURE DERIVATIVE.
pD(der) = t(dpD/dt) qD(der) = t(dqD/dt)
1/pD(der) = ( t(dpD/dt) ) -1
Time Typecurves
Comparison to Blasingame Typecurves
Rate Integral-
Derivative
Inv. Pressure
Integral-
Derivative
qDd and tDd

plotting format
qD and tDA
plotting fomat
Cumulative Typecurves
Agarwal and Gardner Rate vs. Cumulative typecurves are different from
conventional typecurves because they are plotted on LINEAR
coordinates.
They are designed to analyze BOUNDARY DOMINATED data only. Thus,

they do not yield estimates of permeability and skin, only fluid-in-place.
Plot: qD (1/pD) vs QDA
Where (for oil):
141.2B qt 
qD 
kh pi  pwf t 
1 Q 1 pi  p
QDA  qD * tDA  or alternativ ely
2 ctN ( pi  pwf ) 2 pi  pwf
Where (for gas):
1.417e6 * T qt 
qD 
kh  i  wf t 
1 2qtca 1  i 
QDA  qD * tDA  or alternativ ely
2 ctZ iGi ( i  wf ) 2  i  wf
qD vs QDA typecurves
always converge to 1/2
0.159)
NPI (Normalized Pressure Integral)
NPI analysis plots a normalized PRESSURE rather than a normalized
RATE. The analysis consists of three sets of typecurves:
1. Normalized pressure vs. tc (material balance time)
2. Pressure integral vs. tc
3. Pressure integral - derivative vs. tc
- Pressure integral methodology was developed by Tom Blasingame;

originally used to interpret drawdown data with a lot of noise. (ie.
conventional pressure derivative contains far too much scatter)
- NPI utilizes a PRESSRE that is normalized using the current RATE.

It also utilizes the concepts of material balance time and pseudo-
time.
NPI (Normalized Pressure Integral):
Definitions

Normalized Pressure khDP DP DPp
PD 
141.2q q q
dPD  DP  d DPp 
PDd  d  
Conventional d ln t DA   DP   DPp 
   
q q
Pressure Derivative    
 q  d d ln t c   q  i d ln t ca 
 DP  1 c DP  DPp DPp
t DA t

t ca
 P t dt
1 1
PDi     

Pressure Integral
p dt    dt
t DA 0  q i t c 0 q  q  i t ca 0
q
PDid  t DA
dPDi  DP   DPp 
d   t ca d  
Pressure Integral - dt DA  DP   q i  DPp   q i
Derivative    tc   
 q  id dt c  q  id dt ca
Diagnostics
Transient
Normalized
Pressure
Typecruve
Integral - Derivative
Typecurve
Boundary
Dominated
Calculation of Parameters- Oil
Oil - Radial
khDP 0.00634ktc
PD  t DA 
141.2q C t re2
 
 
141.2   PD 
k
h  DP 
 
 q  match
0.00634k  tc 
re   
Ct  t DA  match
re r 
rwq  S  ln  w 
 re   rwa 
 
 rwa  match
 
 
0.00634  141.2 S 0  PD   tc 
N     (MBBIS)
Ct  5.615 * 1000  DP   t DA  match
 
 q  match
Calculation of Parameters- Gas
Gas – Radial
khDPp 0.00634ktca
PD  t DA 
1.4176Tq i C ti re2
 
 
1.4176T  PD 
k  DPp 
h
 
 q  match
0.00634k  t ca 
re   
i Cti  t DA  match
re r 
rwa  S  ln  w 
 re   rwa 
 
 rwa  match
 
 
0.006341.4176S g PiTsc  t ca   PD 
G     * 10 9 (bcf)
 i cti z i Psc  t DA  match  DPp 
 q
  match
Transient (tD format) Typecurves
Transient typecurves plot a normalized rate against material balance time
(similar to other methods), but use a dimensionless time based on
WELLBORE RADIUS (welltest definition of dimensionless time), rather
than AREA. The analysis consists of two sets of typecurves:
1. Normalized rate vs. tc (material balance time)
2. Inverse pressure integral - derivative vs. tc
- Transient typecurves are designed for analyzing EARLY-TIME data to

estimate PERMEABILITY and SKIN. They should not be used (on their
own) for estimating fluid-in-place
- Because of the tD format, the typecurves blend together in the early-time

and diverge during boundary dominated flow (opposite of tDA and tDd
format typecurves)
Transient versus Boundary
Scaling Formats
log(qD)
log(qDd)
log(tD) log(tDd)
Transient (tD format) Typecurves:
Definitions

141.2q q q
Normalized Rate qD 
khDP DP DPp
1 1 1
 1 t DA   DP   1 c DP   DPp   1 DPp 
t t ca
Inverse Pressure
Integral
1 / PDi   P 
 p 
t dt Inv     dt  Inv     dt 
 t DA 0   q i  tc 0 q   q i  tca 0
q 
1 1 1
 dP    DP     DPp  
Inverse Presssure 1 / PDid  t DA Di   d     tca d   
 dt DA   DP   q i   DPp    q i 
 tc
Integral - Derivative
Inv  Inv  
 q id  dtc   q id  dtca 
   
   
Diagnostics (Radial Model)
Transient Transition to Boundary

Dominated occurs at
Inverse Integral -
Derivative
different points for
Typecurve different typecurves
Normalized Rate
Typecurve
Finite Conductivity Fracture Model
Increasing Fracture
Conductivity (FCD
stems)
Increasing
Reservoir Size
(xe/xf stems)
Calculations (Radial Model)
Oil Wells:
Gas Wells:
Using the definition of qD,
For gas wells, qD is defined as follows:
141.2qB
qD 
kh( pi  pwf )
1.417 E 6TR q
qD 
kh Dpp
permeability is calculated as follows:
141.2 B  q/Dp  The permeability is calculated from above, as follows:

k  qD 
h   match
1.417 E 6TR  q/Dpp 

k  
From the definition of tD, h  qD  match
0.00634ktc
tD 
ctrwa 2
From the definition of tD and k, rwa is calculated as follows
rwa is calculated as follows: 0.00634  1.417 E 6TR  tca   q/Dpp 

rwa      
icti  h   match  qD  match
t D
0.00634  141.2 B   q/Dp   tc 

rwa       
ct  h   qD  match  tD  match Skin is calculated as follows:
Skin is calculated as follows:  rw 

s  ln 
 rwa 
 rw 
s  ln 
 rwa 
Flowing p/z Method for Gas –
Constant Rate
- Mattar L., McNeil, R., "The 'Flowing' Gas
Material Balance", JCPT, Volume 37 #2, 1998
pi
zi Pressure loss due to flow
in reservoir (Darcy’s Law)
is constant with time
pwf
p  p
zwf     constant
z  z  wf
Gi
Measured at well
during flow
Gp
Graphical Flowing p/z Method
for Gas – Variable Rate
pi
zi
Graphical Method Doesn’t
Work!
pwf
zwf
Gi ?
Measured at well
during flow
Gp
Flowing p/z Method for Gas –
Variable Rate
pi
zi Pressure loss due to flow
in reservoir is NOT
constant
pwf
zwf p  p
    qbpss
z  z  wf
Unknown
Gi
Measured at well
during flow
Gp
Variable Rate p/z – Procedure (1)
Legend
550
Static P/Z *
P/Z Line
Flow ing Pressure
500
450
Step 1: Estimate OGIP and

400
plot a straight line from pi/zi
to OGIP. Include flowing
Flowing Pressure,
350
pressures (p/z)wf on plot 300
250
psi
200
150
100
Original Gas In Place 50
0
0.00 0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 0.90 1.00 1.10 1.20 1.30 1.40 1.50 1.60 1.70 1.80 1.90 2.00 2.10 2.20 2.30 2.40 2.50 2.60 2.70
Cum ulative Production, Bscf

Legend
550
Static P/Z *
4.40 P/Z Line
Flow ing Pressure
500
Productivity Index
4.00
3.60
Step 2: Calculate bpss for 450
each production point using

6 psi 2 /cP)
400
3.20
the following formula:
Flowing Pressure,
350
MMscfd/(10
2.80
 p  p 300
2.40
    
 z line  z  wf
Productivity Index,
250
bpss 
2.00
psi
1.60 q 200
150
1.20
Plot 1/bpss as a function of
0.80 Gp 100
0.40 Original Gas In Place 50
0.00 0
0.00 0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 0.90 1.00 1.10 1.20 1.30 1.40 1.50 1.60 1.70 1.80 1.90 2.00 2.10 2.20 2.30 2.40 2.50 2.60 2.70

Legend
550
Static P/Z *
4.40 P/Z Line
Flow ing Pressure
500
Productivity Index
4.00
Step 3: 1/bpss should tend 450

3.60
towards a flat line. Iterate on

6 psi 2 /cP)
400
3.20
OGIP estimates until this
Flowing Pressure,
350
happens
MMscfd/(10
2.80
300
2.40
Productivity Index,
250
2.00
psi
200
1.60
150
1.20
0.80 100
0.40 50
0.00 0
0.00 0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 0.90 1.00 1.10 1.20 1.30 1.40 1.50 1.60 1.70 1.80 1.90 2.00 2.10 2.20 2.30 2.40 2.50 2.60 2.70

Legend
550
Static P/Z *
4.40 P/Z Line
Flow ing P/Z *
500
Flow ing Pressure
4.00 Productivity Index
450
3.60
Step 4: Plot p/z points on the
p/z line using the following
6 psi 2 /cP)
400
3.20
P/Z * , Flowing Pressure,

formula: 350
MMscfd/(10
2.80
 p  p
2.40
      qbpss
300
 data  z  wf
z
Productivity Index,
250
2.00
psi
1.60 “Fine tune” the OGIP estimate 200
150
1.20
0.80 100
0.40
1/bpss 50
0.00 0
0.00 0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 0.90 1.00 1.10 1.20 1.30 1.40 1.50 1.60 1.70 1.80 1.90 2.00 2.10 2.20 2.30 2.40 2.50 2.60 2.70

Specialized
Modeling and History Matching
1. Pressure Constrained System:
Constraint (Input) Signal (Output)

Well / Reservoir
Well Pressure at Sandface Model Production Volumes
2. Rate Constrained System:
Constraint (Input) Well / Reservoir Signal (Output)

Production Volumes Model Well Pressure at Sandface
Models - Radial
Rectangular reservoir with a vertical well located anywhere inside.
Models - Horizontal
Rectangular reservoir with a horizontal well located anywhere inside.
Models - Fracture
Rectangular reservoir with a vertical infinite conductivity fracture located anywhere inside.
A Systematic and Comprehensive
Method for Analysis
Modern Production Analysis
Methodology
Diagnostics Interpretation and Modeling and Forecasting

Analysis History Matching
- Data Validation - Identifying dominant - Validating interpretation

flow regimes - Reserves
- Reservoir signal - Optimizing solution
- Estimating reservoir - Optimization scenarios
extraction - Enabling additional
characteristics flexibility and complexity
- Identifying important
system parameters
- Qualifying
uncertainty
- Data Chart - Traditional - Analytical Models

- Typecurves - Fetkovich - Numerical Models
- Blasingame
- AG / NPI
- Flowing p/z
- Transient
Practical Diagnostics
What are diagnostics?
• Qualitative investigation of data

– Pre-analysis, pre-modeling
– Must be quick and simple
• A VITAL component of production data

analysis (and reservoir engineering in
general)
Illustration- Typical Dataset
28
Legend 1600
5.50
26 Pressure 1500
Actual Gas Data
24 5.00 1400
1300
22
4.50
1200
20
4.00 1100
18
1000
Liquid Rates , bbl/d
3.50
Pressure , psi
16
G as , MMcfd
900
14 800
3.00
12 700
2.50 600
10
500
8 2.00
400
6
1.50 300
4
200
1.00
2 100
0 0.50 0
0 20 40 60 80 100 120 140 160 180 200 220 240 260 280 300 320 340 360 380 400 420 440 460 480 500 520 540
Tim e, days
“Face Value” Analysis of Data
OGIP = 90 bcf
Go Back: Diagnostics
28
Legend 1600
5.50
26 Pressure 1500
Actual Gas Data
24 5.00 1400
1300
22
4.50
1200
20
4.00 1100
18
1000
3.50
Pressure , psi
16
G as , MMcfd
900
14 Legend
800
3.00 Pressure
Actual Gas Data
12 700
2.50 600
10
500
8 2.00
Pressures are not 400
6
1.50
representative of 300
4
2
1.00 bh deliverability 200
100
0 0.50 0
0 20 40 60 80 100 120 140 160 180 200 220 240 260 280 300 320 340 360 380 400 420 440 460 480 500 520 540
Tim e, days
Correct Data Used
Unnamed Well Data Chart
7400
Legend
6.00 5.50
Pressure 7200
Actual Gas Data
5.50 Oil Production
5.00 7000
Water Production
5.00 6800
4.50
4.50 6600
4.00
4.00 6400
3.50 6200
Pressure , psi
Gas , MMcfd
3.50
6000
3.00 3.00
5800
2.50
2.50
5600
2.00
2.00 5400
OGIP = 19 bcf
1.50
5200
1.50
1.00
5000
1.00
0.50 4800
0.00 0.50 4600

0 20 40 60 80 100 120 140 160 180 200 220 240 260 280 300 320 340 360 380 400 420 440 460 480 500 520 540
Time, days
Diagnostics using Typecurves
Radial Model
Blasingam e Typecurve Match
10-7
8
5
3
2
10-8
8
5
Transient
qDd 3 (concave up) Boundary Dominated
2
(concave down)
10-9
8
5
3
2
Base Model:
10-10
5
8
- Vertical Well in Center of Circle
3
2
- Homogeneous, Single Layer
10-11
8
5
3
2
4 56 8 -1
10
2 3 45 79
1.0
2 3 45 7 9 1
10
2 3 4 56 8 2
10 tDd2
2 3 4 56 8 3
10
3 4 56 8 4
10
2 3 4 56 8 5
10
2 3 4 56 8 6
10
2 3 45 7
107
2 3 45 7
Material Balance Diagnostics
Radial Model
10-7
8
5
3
2 Reservoir With
10-8
8
Pressure Support
5
qDd 3
2
10-9
8
5
3
2
10-10
8
5
3
2
10-11
8
Leaky Reservoir
5
3 (interference)
2
4 56 8 -1
10
2 3 45 79
1.0
2 3 45 7 9 1
10
2 3 4 56 8 2
10 tDd2
2 3 4 56 8 3
10
3 4 56 8 4
10
2 3 4 56 8 5
10
2 3 4 56 8 6
10
2 3 45 7
107
2 3 45 7
Productivity Diagnostics
Radial Model
10-7
8 Increasing Damage (difficult to identify)
5
3
2
10-8
8
5
qDd 3
Productivity Shifts
2 (workover,
10-9
8
unreported tubing
5
Well Cleaning Up change)
3
2
10-10
8
Liquid Loading
5
3
2
10-11
8
5
3
2
4 56 8 -1 2 3 45 79 2 3 45 7 9 1 2 3 4 56 8 2 2 3 4 56 8 3
10 1.0 10 10 tDd 2
10
3 4 56 8 4
10
2 3 4 56 8 5
10
2 3 4 56 8 6
10
2 3 45 7
107
2 3 45 7
Transient Flow Diagnostics
Radial Model
10-7 Fracture Linear Flow

8
5 (Stimulated)
3
Transitionally
2 Dominated Flow (eg:
10-8
8
Channel or Naturally
5 Damaged Fractured)
qDd 3
2
10-9
Radial Flow
8
5
3
2
10-10
8
5
3
2
10-11
8
5
3
2
4 56 8 -1 2 3 45 79 2 3 45 7 9 1 2 3 4 56 8 2 2 3 4 56 8 3 2 3 4 56 8 4 2 3 4 56 8 5 2 3 4 56 8 6 2 3 45 7 2 3 45 7
10 1.0 10 10 tDd
10 10 10 10 107
“Bad Data” Diagnostics
Radial Model
10-7
8 Dp in reservoir is too low
5
-Tubing size too small ?
3
2 - Initial pressure too low ?
10-8
- Wellbore correlations
8
overestimate pressure loss ?
5
qDd 3
2
10-9
8
5 Dp in reservoir is too high
3
2
-Tubing size too large ?
- Initial pressure too high ?
10-10
8 - Wellbore correlations
5
underestimate pressure loss ?
3
2
10-11
8
5
3
2
4 56 8 -1
10
2 3 45 79
1.0
2 3 45 7 9 1
10
2 3 4 56 8 2
10 tDd 2
2 3 4 56 8 3
10
3 4 56 8 4
10
2 3 4 56 8 5
10
2 3 4 56 8 6
10
2 3 45 7
107
2 3 45 7
Selected Topics and Examples
Tight Gas
Industry Migration to Tight Gas
Reservoirs
Production Analysis – Tight Gas versus
Conventional Gas
 Analysis methods are no different from that
of high permeability reservoirs
 Transient effects tend to be more dominant

– Establishing the region (volume) of
influence is critical
 Drainage shape becomes more important

(Transitional effects)
 Linear flow is more common
 Layer effects are more common

Tight Gas- Common Geometries
Tight Gas Type Curves
1.00E-05
Infinite acting reservoir

1.00E-06
1/2
1.00E-07
qDd
1.00E-08
1
1.00E-09
Linear flow
1.00E-10
dominated Limited, bounded
drainage area
1.00E-11
1.00E-05 1.00E-04 1.00E-03 1.00E-02 1.00E-01 1.00E+00 1.00E+01 1.00E+02 1.00E+03 1.00E+04
tDd
Tight Gas Model 1
 Extensive, continuous porous media; very low

permeability
1800 psi
Pi = 2000 psi
Pi = 1500 psi
Infinite Acting System
1.00E-05
1.00E-06
1.00E-07
1/2
qDd
1.00E-08
1.00E-09
1.00E-10
1.00E-11
1.00E-05 1.00E-04 1.00E-03 1.00E-02 1.00E-01 1.00E+00 1.00E+01 1.00E+02 1.00E+03 1.00E+04
tDd
Example#1 – Infinite Acting System
10 Agarwal Gardner Rate vs Time Typecurve Analysis 10 Agarwal Gardner Rate vs Time Typecurve Analysis
2 2
102 102
6 6
4 4
3 3
2 2
101 101
7 7
5 5
3 3
2 2
Normalized Rate
Normalized Rate
1.0 1.0
9 9
6 6
4 4
3 3
2 2
10-1 10-1
7
7
5
5
3
3
2
2
10-2
10-2
7
7
5
5
3
3
2
2
2 3 4 5 6 78 2 3 4 5 6 78 2 3 4 5 6 78 2 3 4 5 6 78 2 3 4 5 6 78 2 3 4 5 6 78 2 3 4 5 6 78
10-5 10-4 10-3 10-2 10-1 1.0 101 102 2 3 4 5 6 78 2 3 4 5 6 78 2 3 4 5 6 78 2 3 4 5 6 78 2 3 4 5 6 78 2 3 4 5 6 78 2 3 4 5 6 78
10-5 10-4 10-3 10-2 10-1 1.0 101 102
Material Balance Pseudo Tim e
k = 0.08 md k = 0.08 md
xf = 53 ft xf = 53 ft
OGIP = 10 bcf Minimum OGIP = 2.6 bcf
Tight Gas Model 2
 No flow continuity across reservoir- Well only

drains a limited bounded volume
Example: Lenticular Sands

Bounded Reservoir
1.00E-05
1.00E-06
1.00E-07
1/2
qDd
1.00E-08
1.00E-09
- Limited or no flow continuity in reservoir
1
- Very small drainage areas
- Very large effective fracture lengths
1.00E-10
1.00E-11
1.00E-05 1.00E-04 1.00E-03 1.00E-02 1.00E-01 1.00E+00 1.00E+01 1.00E+02 1.00E+03 1.00E+04
tDd
Commonly observed in practice

Example #2- Bounded Drainage
Areas
ROBINSON 11-1 ALT
2
10
101
- West Louisiana gas field

7
5
9 3
- 80 acre average spacing 2
1.0
- All wells in boundary dominated flow
Normalized Rate
7
8 5
10-1
7 7
5
3
.
6
10-2
2 3 4 5 6 78 2 3 4 5 6 78 2 3 4 5 6 78 2 3 4 5 6 78 2 3 4 5 6 78
10-3 10-2 10-1 1.0 101 102

OGIP (bcf)
35 120%
4
30
100%
.
3 25
80%
Frequency
20
60%
2
15
40%
10
1
20%
5
0 0 0%
10 20 30 40 50 60 70 80 90 100 More
0 100 200 300 400 500 600
Drainage Area (acres)
xf (feet)
Frequency Cumulative %
Tight Gas Model 3
 Linear flow dominated system
Example: Naturally fractured, tight reservoir
kx
ky
Infinite Systems versus Linear Flow
Systems
Establish
permeability and
xf independently
Establish xf sqrt
(k) product only
Linear Flow Systems
1.00E-05
1.00E-06
1.00E-07
1/2
qDd
1.00E-08
- Channel and faulted reservoirs

1.00E-09
- Naturally fractured (anisotropic) reservoirs
- Very large effective fracture lengths
1.00E-10 - Very difficult to uniquely interpret
1.00E-11
1.00E-05 1.00E-04 1.00E-03 1.00E-02 1.00E-01 1.00E+00 1.00E+01 1.00E+02 1.00E+03 1.00E+04
tDd
Commonly observed in practice

Example #3- Linear Flow System
Fracture Model
5
4
3
k = 1.1 md
xf = 511 ft
2
ye = 5,500 ft
yw = 2,900 ft
10-7
ye
7
yw
5
4
2xf
10-8
9
2 3 4 5 6 7 8 9 2 3 4 5 6 7 8 9 2 3 4 5 6 7 8
101 102 103
More Examples
Example #3- Multiple Layers
Multi Layer Model

3
Well Blasingam e Typecurve Match
10-8
2
5
4
1.0 3
8
2
7
Normalized Rate
5
10-9
4 8
6
3
3
2
10-1
10-10
9
2 3 4 5 6 7 89 2 3 4 5 6 7 89 2 3 4 5 6 7 89 2 3 4 5 6 78
1.0 101 102 103 104
3 4 5 6 7 8 9 2 3 4 5 6 7 8 9 2 3 4 5 6 7 8 9
10-1 1.0
- Blasingame typecurve match, using Fracture Model - Three-Layer Model (one layer with very low
- Pressure support indicated permeability) used, late-time match improved
Example #4- Shale Gas
Well Agarwal Gardner Rate vs Time Typecurve Analysis
3
- Multi-stage fractures, horizontal well
- Analyzed as a vertical well in a circle
2
1.0
7
6
5
Normalized Rate
10-1
9
7
6
5
k = 0.02 md
4 s = -4
3
OGIP = 4.5 bcf
6 7 8 9 2 3 4 5 6 7 8 9 2 3 4 5 6 7 8 9 2 3 4 5 6 7 8 9
10-3 10-2 10-1 1.0

Tight Gas: Assessing Reserve Potential
– Recovery Plots
 Objectives
 Determine incremental reserves that are added as the

ROI expands into the reservoir (only relevant for
infinite or semi-infinite systems)
 To establish a practical range of Expected Ultimate

Recovery
Typical Recovery Profile
Recovery Curves for k = 1 md
10
9
1 md reservoir, unfractured
8 (~10 bcf / section)
7 100% Recovery
6
EUR (bcf)
0
0 1 2 3 4 5 6 7 8 9 10
Original Gas in Place (bcf)
10
9
1 md reservoir, unfractured
7 100% Recovery
6
EUR (bcf)
3
Actual EUR (qab = 0.05 MMscfd)
2
0
0 1 2 3 4 5 6 7 8 9 10
EUR- unlimited time

10
9 1 md reservoir, unfractured
8
(~10 bcf / section)
7 100% Recovery
6
30 Year Limited
EUR (bcf)
3
2
0
0 1 2 3 4 5 6 7 8 9 10
EUR- 30 year EUR- unlimited time

10
9 1 md reservoir, unfractured
8
(~10 bcf / section)
7
100% Recovery
6
EUR (bcf)
5 30 Year Limited
4
3 20 Year Limited
2
1
0
0 1 2 3 4 5 6 7 8 9 10
EUR- 30 year EUR- 20 year EUR- unlimited time

Tight Gas Recovery Profile
Recovery Curves for k = 0.02 md
10
9 0.02 md reservoir,
fractured
8
(~10 bcf / section)
7
6
EUR (bcf)
0
0 1 2 3 4 5 6 7 8 9 10
EUR- unlimited time

10
9
0.02 md reservoir,
fractured
8
(~10 bcf / section)
7
6
EUR (bcf)
4
30 Year
3
0
0 1 2 3 4 5 6 7 8 9 10
EUR- 30 year EUR- unlimited time

10
9
0.02 md reservoir, fractured
7 Actual EUR (qab = 0.05 MMscfd)
6
EUR (bcf)
4
30 Year
3 20 Year
2
0
0 1 2 3 4 5 6 7 8 9 10

10
9
0.02 md reservoir,
fractured
8
(~10 bcf / section)
7 Actual EUR (qab = 0.05 MMscfd)
6 Max EUR (30 y) = 2 bcf
EUR (bcf)
4
30 Year
3 20 Year
2
0
0 1 2 3 4 5 6 7 8 9 10

Example – South Texas, Deep
Gas Well
Fracture Model
AG Typecurve Match
10-8
5
4
Sqrt k X xf = 155
3 Min OGIP = 4.2 bcf
2
10-9
9
7
2 3 4 5 6 7 8 9 2 3 4 5 6 7 89 2 3 4 5 6 7 89 2 3 4 5 6 7 8
1.0 101 102 103
Example – South Texas, Deep
Gas Well
Recovery Plot - Linear System
Maximum EUR = 6.7 bcf

6
Recovery period = 30 years

sqrt k X xf = 155
4
EUR (bcf)
pi = 6971 psia
Minimum EUR = 3.5 bcf
0
0 100 200 300 400 500 600
ROI (acres)
Water Drive Models
Water Drive (Aquifer) Models:
Models for reservoirs under the influence of active water encroachment can
be categorized as follows:
1. Steady State Models (inaccurate for finite reservoir sizes)

- Schilthuis
2. Pseudo Steady-State Models (geometry independent,

time discretized)
- Fetkovich
3. Single Phase Transient Models (geometry dependent)

- infinite aquifer (linear, radial or layer geometry)
- finite aquifer (linear, radial or layer geometry)
4. Modified Transient Models

- Moving saturation front approximations
- Two phase flow approximations
Pseudo Steady-State Models
PSS models (such as that of Fetkovich) use a TRANSFER
COEFFICIENT (similar to a well productivity index) to describe the
PSS rate of water influx into the reservoir, in conjunction with a
MATERIAL BALANCE model that predicts the decline in reservoir
boundary pressure over time.
The Fetkovich model is generally used to determine reservoir fluid-

in-place by history matching the CUMULATIVE PRODUCTION and
AVERAGE RESERVOIR PRESSURE.
Pseudo Steady-State Models
Advantages:
- Geometry independent (applicable to aquifers of any shape, size or

connectivity to the reservoir)
- Works well for finite sized aquifers of medium to high mobility
- Computationally efficient
Disadvantages:
- Does not provide a full time solution (transient effects are ignored)
- Does not work well for infinite acting or very low mobility aquifers
Pseudo Steady-State Model- Equations
The Fetkovich water influx equation for a finite aquifer is:
Aquifer transfer coefficient
We 
Wei
 pi-p  1  e Jpit /Wei  Initial encroachable water
pi  
Reservoir boundary pressure
The above equation applies to the water influx due to a constant pressure difference between aquifer and
reservoir. In practice, the reservoir pressure “p” will be declining with time. Thus, the equation must be
discretized as follows:
DWe n 
Wei
pi
 
pa n1- p n 1  e

 Jpit /Wei 


(1)
The average aquifer pressure at the previous timestep (n-1) is evaluated explicitly, as follows:
 n 1 
  DWej 
 pi 1  
j 1
pa n1
Wei 
 
 
Pseudo Steady-State Model- Equations
Now, we have one equation with two unknowns (water influx “We” and reservoir boundary
pressure “p”)
But there is another equation that relates the average reservoir pressure to the amount of water
influx: the material balance equation for a gas reservoir under water drive.
Cumulative Production
-1
p pi  Gp   WeBi 
 1   1   FVF at initial conditions
z zi  Gi   Gi  Gas-in-place
As with the water influx equation, the material balance equation can be discretized in time:
-1
 p pi  Gp n   We n Bi 
   1   1   (2)
 z  n zi  Gi   Gi 
Equations 1 and 2 are now solved simultaneously at each timestep, to obtain a discretized
reservoir pressure and water influx profile through time.
Transient Models
Transient models use the full solution to the hydraulic DIFFUSIVITY EQUATION to
model rates and pressures.
The transient equations can be used to model either FINITE or INFINITE acting
aquifers. There are a number of different transient models available for analyzing
a reservoir under active water drive:
- Radial Composite (edge water drive)

- Linear (edge water drive)
- Layered (bottom water drive)
Advantages:
- Offers full continuous pressure solution in the reservoir

- Includes early time effects
Disadvantages:
- Geometry dependent (only a disadvantage if aquifer properties are unknown)

- Limited to assumption of single phase flow
- Does not account for water influx
Water Drive (Aquifer) Typecurves:
Radial Composite Model
Blasingame, AG and NPI dimensionless formats can be used to plot
typecurves for SINGLE PHASE production (oil or gas) from a reservoir under
the influence of an EDGE WATER DRIVE. A typecurve match using this
model can be used to predict
1. Reservoir fluid-in-place
2. Aquifer mobility
- These typecurves are designed to estimate fluid-in-place by

detecting the shift in fluid mobility as the transient passes the reservoir
boundaries, into the aquifer.
- Their usefulness is limited to single phase flow (ie: the transition from
reservoir fluid to aquifer is assumed to be abrupt)
Definitions
Model Type: Radial Composite (two zones);
outer zone is of infinite extent
Reservoir Aquifer
Mobility Ratio (M):
Maq kaq res

M 
Mres kres aq
Diagnostics
M=10 (Constant Pressure System

(approx))
Decreasing reD value
Increasing Aquifer Mobility

(M)
M=0 (Volumetric Depletion)

Diagnostics
M=10 (Constant Pressure System

(approx))
Increasing Aquifer Mobility

(M)
Decreasing reD value
M=0 (Volumetric Depletion)

Modified Transient Models
1. Moving aquifer front (reservoir boundary)
The radial composite model previously discussed can be enhanced to

accommodate a shrinking reservoir boundary, caused by water influx.
This is achieved by discretizing the transient solution in time and using
the PSS water influx equations to predict the advancement of the aquifer
front. The solution still assumes single phase flow, but can now more
accurately estimate the time to water breakthrough.
2. Two phase flow (after M. Abbaszadeh et al)
The previously discussed model can also be modified to accommodate a

region of two-phase flow (located between the inner region - hydrocarbon
phase and outer region - water phase). Thus, geometrically, the overall
model is three zone composite. The pressure transient solution for the
two-phase zone is calculated by superimposing the single phase
pressure solution on a saturation profile determined using the Buckley-
Leverett equations.
Water Drive (Aquifer) Models: Example
Exam ple F Data Chart Exam ple F Blasingame Typecurve Analysis
14000 101
Legend 8
22 Pressure
Actual Gas Data 13000 6
5
20 4
12000
18 11000
2
10000
16
1.0
9000
8
14
6
8000 5
Pressure,
Normalized Rate
Gas, MMscfd
4
12
7000
3
psi
10 6000 2
5000
8
6
4000
10-1
8
6
-Boundary dominated
4
-Gulf coast gas 3000
5
3
-Pressure support evident
2000
2 condensate reservoir 1000

2
0 0 10-2
Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep Oct 2 3 4 5 6 7 8 2 3 4 5 6 7 8 2 3 4 5 6 7 8 2 3 4 5 6 7 8 2 3 4 5 6 7 8
10-3 10-2 10-1 1.0 101 102
2002 2003
Exam ple F Agarwal Gardner Rate vs Time Typecurve Analysis Exam ple F Blasingame Typecurve Analysis
101
8
1.0
8
Transient Water Drive 6
5
4 PSS Water Drive Model
6
5
4
Model 3
2
3
2
1.0
6
Normalized Rate, Derivative
10-1 5
Normalized Rate
6
5
4
3
k = 3.1 md
4
3 k = 8.5 md
2
s = -4
2
s=0
10-1
8 OGIP = 13.5 bcf
6
10-2
8 OGIP = 12 bcf
5
4 IWIP = 47 MMbbl
3
6
5
4
M = 0.001 2 PI (aq) = 0.59 bbl/d/psi
3
10-2
2 3 4 5 6 7 8 9 2 3 4 5 6 7 8 9 2 3 4 5 6 7 8 9 2 3 4 5 6 7 8 2 3 4 2 3 4 5 6 7 8 2 3 4 5 6 7 8 2 3 4 5 6 7 8 2 3 4 5 6 7 8 2 3 4 5 6 7 8
10-1 1.0 101 102 10-3 10-2 10-1 1.0 101 102
Material Balance Pseudo Tim e Material Balance Pseudo Tim e

Multiple Well Analysis
Multi-well / Reservoir-based Analysis-
Available Methods
1. Empirical- Group production decline plots
2. Material Balance Analysis- Shut-in data only
3. Reservoir Simulation
4. Semi-analytic production data analysis methods

- Blasingame approach
Multi-Well Analysis- When is it
required?
1. Situations where high efficiency is required
- Scoping studies / A & D
- Reserves auditing
2. Single well methods sometimes don’t apply

- Interference effects evident in production / pressure
data- Wells producing and shutting in at different times
- Predictive tool for entire reservoir is required
- Complex reservoir behavior in the presence of
multiple wells (multi-phase flow, reservoir
heterogeneities)
Multi-Well Analysis- When is it not
required?
The vast majority of production data can be analyzed
effectively without using multi-well methods
1. Single well reservoirs
2. Low permeability reservoirs

- Pressure transients from different wells in reservoir
do not interfere over the production life of the well
3. Cases where “outer boundary conditions” do not change

too much over the production life of the well
- Wide range of reservoir types
Identifying Interference
Well A Well B
Rate is adjusted at Well A Response at Well B
Q Q
Correcting Interference Using
Blasingame et al Method
Define a “total material balance time” function
Qtot QA  QB
tce   (for analyzing Well A)
q qA
tce is used in place of tc to plot the data in the typecurve match

Multi-Well Analysis as a
Typecurve Plot
MBT is corrected for
Analysis of Well A: interference caused
by production from
Well B
log(q/Dp)
log(tc) tcA tce
tce= (QB +QA)/qA
Also applies to Agarwal-Gardner, NPI and FMB

Multi-Well Analysis- Example
Well 1 Data Chart
6.00 36000
Legend
2.80
Pressure 34000
5.50
2.60 -Three well system Actual Gas Data
Pool Production 32000
5.00 2.40
-“Staggered” on-stream dates Water Production
30000
-High permeability reservoir 28000

2.20
4.50
26000
2.00
24000
4.00
Aggregate production of well group
1.80
bbl/d
22000
3.50
Pressure,
Gas, MMscfd
1.60 20000
Oil / Water Rates,
3.00 18000
1.40
psi
16000
2.50 1.20
14000
1.00
2.00 12000
0.80 10000
1.50
8000
0.60
1.00 6000
0.40
4000
0.50 0.20 Production history of well to be analyzed 2000
0.00 0.00 0
1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002
Well 1 Blasingame Typecurve Analysis
101
5
4
Well 1 Blasingame Typecurve Analysis

101
1.0
7
7
5
5 4
Normalized Rate
4
3
3
2
2
1.0
10-1
7
7
“Leaky reservoir” diagnostic 5
Normalized Rate
5
4
4
3
3
2
2
10-1
10-2
2 3 4 5 6 78 2 3 4 5 6 78 2 3 4 5 6 78 2 3 4 5 6 78 2 3 4 5 6 78 7
10-3 10-2 10-1 1.0 101 102
Material Balance Pseudo Tim e 5 Corrected using multi-well model

4
3 Total OGIP = 7 bcf

2
10-2
2 3 4 5 6 78 2 3 4 5 6 78 2 3 4 5 6 78 2 3 4 5 6 78 2 3 4 5 6 78
10-3 10-2 10-1 1.0 101 102

Well 1 Flowing Material Balance
Legend 1900
P/Z Line
Flow ing P/Z * 1800
1700
1600
1500
1400
OGIP for subject well = 3.5 bcf 1300
1200
Well 1 Flowing Material Balance
1100
Legend
P/Z
2000
1000 P/Z Line
*,
psi
Flow ing P/Z *
900
1800
800
700
1600
600 Total OGIP = 7.0 bcf
500
1400
400
300 1200
200
P/Z
*,
100 1000
psi
0
0.00 0.40 0.80 1.20 1.60 2.00 2.40 2.80 3.20 3.60 4.00 4.40 4.80 5.20 5.60 6.00 6.40 6.80 7.20 7.60
800
600
400
200
0
0.00 0.40 0.80 1.20 1.60 2.00 2.40 2.80 3.20 3.60 4.00 4.40 4.80 5.20 5.60 6.00 6.40 6.80 7.20 7.60 8.00

Overpressured Reservoirs
Overpressured Reservoirs
1. Analysis methods are the same as normally pressured case
2. Additional parameters to be aware of

• Formation compressibility
• In-situ water compressibility
• Compaction effects (pressure dependent permeability)
3. Two models available, depending on required complexity

• p/z* model (accounts for constant cf, cw and co in
material balance equation
• Full geomechanical model (accounts for cf(p) and k(p))
Compresibilities of Gas and Rock
Compressibility vs. Pressure (Typical Gas Reservoir)
3.00E-04
gas
Formation Formation
2.50E-04 energy is energy may Formation energy is critical in this region
negligible in be influencial
Compressibility (1/psi)
this region in this region

2.00E-04
1.50E-04
1.00E-04
5.00E-05 formation
0.00E+00
0 2000 4000 6000 8000 10000 12000
Reservoir Pressure (psi)
p/z* Model – Corrects Material
Balance
 p 
p 1 Gp 
   1
  

z 1  cf ( pi  p )  z  i  OGIP 
 
*
Flowing MB
p  p   Gp 
   1 
z  z 
i  OGIP 
  ct  i t q(t ) Typecurves
tca 
q 0  ct 1  cf ( pi  p) dt
Geomechanical Model – Corrects
Well Productivity
In the standard pressure transient equations, permeability is usually considered to be
constant. There are several situations where this may not be a valid assumption:
1. Compaction in overpressured reservoirs

2. Very low permeability reservoirs in general
3. Unconsolidated and/or fractured formations
One way to account for a variable permeability over time is to modify the definition of
pseudo-pressure and pseudo-time.
2qpi 1.417e6 * Tq  re 3 
Dpp *  ta *   ln  
( ctZ )iGi kih  rwa 4 
where
2 pi k ( p) pdp Pressure dependent

Dpp   *
permeability included in
ki pwf z pseudo-pressure and pseudo-
time
( ct )i t k dt
 
*
ta
ki 0 c t
Overpressured Reservoirs -
Example
101
8
6
5
4
Gulf Coast, deep gas condensate reservoir
3
1.0
6
5
Normalized Rate
10-1
8
6
5
Boundary dominated flow
4
OGIP = 17 bcf
3
10-2
2 3 4 5 6 7 8 2 3 4 5 6 7 8 2 3 4 5 6 7 8 2 3 4 5 6 7 8 2 3 4 5 6 7 8
10-3 10-2 10-1 1.0 101 102

Example
Radial Model
218 Prod and Pressure Data History Match
80 18000
70 16000
60 14000
Good flowing pressure match,

50 12000
Poor shut-in pressure match
Pressure, psi
Rate, MMscfd
40
OGIP = 17 bcf 10000
8000
30
6000
20
4000
10
2000
0
June July August Septem ber October
2003
Example
Radial Model
218 Prod and Pressure Data History Match
80 18000
70 16000
60 14000
Good flowing pressure match,

50 12000
Good shut-in pressure match
Pressure,
Rate, MMscfd
OGIP = 29 bcf 10000

40
psi
8000
30
6000
20
4000
10
2000
0
June July August Septem ber October
2003
Example
k (p) Permeability
218 Prod and Pressure Data k (p)
1.05
Legend
1.00 Default
Custom
0.95
Interpolation
0.90
0.85
0.80
0.75
0.70
0.65
Assumed permeability profile
0.60
0.55
k / ki
0.50
0.45
0.40
0.35
0.30
0.25
0.20
0.15
0.10
0.05
0.00
0 500 1000 2000 3000 4000 5000 6000 7000 8000 9000 10000 11000 12000 13000 14000 15000 16000 17000 18000
Pressure, psi(a)
Horizontal Wells
Horizontal Wells
Horizontal wells may be analyzed in any of three different
ways, depending on completion and petrophysical details:
1. As a vertical well,
• if lateral length is small compared to drainage area
2. As a fractured well,
• if the formation is very thin
• if the vertical permeability is high
• if the lateral is cased hole with single or multiple stage
fractures
• to get an idea about the contributing lateral length
3. As a horizontal well (Blasingame model)
• all others
Horizontal Wells – Blasingame
Typecurves
The horizontal well typecurve matching procedure is based on a square shaped reservoir with uniform thickness (h).
The well is assumed to penetrate the center of the pay zone.
The procedure for matching horizontal wells is similar to that of vertical wells. However, for horizontal wells, there is
more than one choice of model. Each model presents a suite of typecurves representing a different penetration ratio
(L/2xe) and dimensionless wellbore radius (rwD). The definition of the penetration ratio is illustrated in the following
diagram:
Plan
Cross Section
L
h rwa
L
2xe
The characteristic dimensionless parameter for each suite of horizontal typecurves is defined as follows:
2xe
2rwa
rwD 
L
Where is the square root of the anisotropic ratio:
 For an input value of “L”,
L kh
LD  
2h kv
Horizontal Wells – Example
Unnam ed Well Blasingame Typecurve Analysis
102
8
6
4
3
L/2xe = 1
2 rwD = 2e-3
101
Ld = 5
8
6 Le = 1,968 ft
4
3
2 k (hz) = 0.18 md
Normalized Rate
1.0
k (v) = 0.011 md
8
6
OGIP = 1.1 bcf
4
3
10-1
8
6
4
3
10-2
2 3 4 5 6 7 8 2 3 4 5 6 7 8 2 3 4 5 6 7 8 2 3 4 5 6 7 8 2 3 4 5 6 7 8
10-3 10-2 10-1 1.0 101 102

Oil Wells
Oil Wells
 Analysis methods are no different from that

of gas reservoirs (in fact they are simpler)
provided that the reservoir is above the
bubble point
 If below bubble point, a multi-phase

capable model (Numerical) must be used
Include relative permeability effects
Include variable oil and gas properties
Oil Wells – Example
exam ple7 Data Chart
4000
190
Legend
0.11 Pressure 3800
Actual Gas Data
180
170 0.10
- Pumping oil well Oil Production
Water Production
3600
3400
160
- Assumed to be pumped off 3200
150
0.09
Producing GOR ~ constant 3000
140 (indicates reservoir pressure is above

2800
130
0.08
bubble point
2600
120
0.07 2400
bbl/d
110
Pressure,
2200
Gas, MMscfd
0.06
Liquid Rates,
100
2000
psi
90
1800
0.05
80
1600
70 1400
0.04
60 1200
50 0.03
1000
40 800
0.02
30 600
20 400
0.01
10 200
0 0.00 0
Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep Oct
2001 2002
Rs input from production data,

Pbp and co calculated using
Vasquez and Beggs
exam ple7 Blasingame Typecurve Analysis
101
8
6
5
4
3
k = 1.4 md
2 s = -3
OOIP = 2.4 million
1.0
bbls
8
6
5
Normalized Rate
10-1
8
6
5
10-2
2 3 4 5 6 7 8 2 3 4 5 6 7 8 2 3 4 5 6 7 8 2 3 4 5 6 7 8 2 3 4 5 6 7 8
10-3 10-2 10-1 1.0 101 102
Material Balance Tim e

exam ple7 Numerical Radial Model - Production Forecast
300 4000
Legend 3800
280 History Oil Rate
240 month forecast Flow Press 3600
Syn Rate
260 EUR = 265 Mbbls History Reservoir Press 3400
Forecasted Press
240 Forecasted Reservoir Press 3200
Forecasted Rate
3000
220
2800
200
2600
180 2400
Pressure,
bbl/d
2200
160
Oil Rate,
2000
psi
140
1800
120 1600
1400
100
1200
80
1000
60 800
600
40
400
20
200
0 0
2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022

Rate Transient Analysis

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Rate Transient Analysis

Uploaded by

Copyright:

Available Formats

Toll Free: 1.800.625.2488 :: Phone: 403.213.4200 :: Email: fast@fekete.

- Using historical trends to predict future

- Empirical (curve fitting)

- Based on physics, not empirical

- Reservoir signal extraction and characterization

- Use BOTH Traditional and Modern together

- Production Data Analysis should include a

- No single method always works

- Production data is varied in frequency, quality

- Graphical – Curve fitting exercise

- Empirical – No theoretical basis

- Implicitly assumes constant operating conditions

Unnam ed Well Rate vs Time

Gas Cum . Prod., Bscf

Gas Cum . Pr od., Bs cf

NBU 921-22G Rate vs. Cumulative Prod.

Gas Cum ulative , Bs cf

- Early-time OR Low Permeability

- Flow that occurs while a pressure “pulse” is

- Like the “fingerprint” of the reservoir

- Late-time flow behavior

- Typically dominates long-term production data

- Reservoir is in a state of pseudo-equilibrium –

- Contains information about reservoir pore volume

Note: only valid if c is constant

pwf2 Constant Rate q

Inflow (Darcy) pressure drop- Constant-

pi  pwf  mpssNp  qbpss

- Take derivative of both equations and solve for q

- Integrate to find Np(t), as t goes to infinity Np goes to

Constant rate q/Dp (Harmonic)

Constant pressure q/Dp (Exponential)

Plan View 800

Plan View 800

Describes radial flow in an infinite acting reservoir

Complete q(t) consists of:

Infinite Acting Boundary Dominated

Constant Rate Solution

- Dimensionless model for reservoir / well system

- Assumes constant operating conditions

- Valuable tool for interpretation of production and

Raw Data Plot Dimensionless Plot

Knowing qi and Di, EUR (expected ultimate recovery) can be calculated

1. Divide the production history into a series of constant rate periods

pi  pwf  q1 f (t )  (q 2  q1) f (t  t1)

- Unit step response f(t) - Real rate and pressure

Superposition Time: Material Balance Time

- MBT converts VARIABLE RATE data into an

- MBT is RIGOROUS for the BOUNDARY

- MBT works very well for transient data also, but

Constant Rate Solution

Constant Pressure Solution qD

Depletion Term Reservoir FlowTerm:

For Gas Flow, this is not true because

Not to be confused with welltest pseudo-time which evaluates properties

Variable Rate Case

0 Single phase liquid expansion (oil above bubble point)

0.1 - 0.4 Solution gas drive

0.4 - 0.5 Single phase gas expansion

0.5 Effective edge water drive

0.5 - 1.0 Layered reservoirs

>1 Transient (Tight Gas)

- Easy and convenient